Table Of ContentsNext Page

Agricultural Farm Analysis and Comparison Tool (AgriFACTs)

Robert M. Aiken1, Vernon L. Thomas2, and William J. Waltman3

1Kansas State University
Northwest Research-Extension Center, 105 Experiment Farm Road, Colby, KS 67701 USA
(785) 462-6281, voice; (785) 462-2315, FAX
raiken@oznet.ksu.edu

2
U.S. Forest Service
Fort Collins, CO 80525
3
University of Nebraska
Lincoln, NE 68901 USA

Abstract

Rainfall frequently limits productivity in the semi-arid regions of the Great Plains. Annual cropping can double water available to crops and land productivity, relative to the traditional wheat-fallow system. Maps of soil and weather may help farmers assess the risks associated with changing cropping systems. Agricultural scientists and GIS specialists integrated results of agroecoregions modeling with cropping systems research to provide farmers with data access. Soil and climatic factors modeled across the central Great Plains landscape included root zone available water holding capacity, precipitation and growing degree days. The climate and soil surface models were developed using terrain regression techniques from point data such as weather station and soil characterization sites, as well as from soil geographic databases. Responses from farmers, crop consultants and agricultural scientists come from workshops, symposia and the Internet. Thematic maps of soil and weather characteristics often grasp and retain a dryland farmer’s attention. Products similar to AgriFACTs can stimulate problem-solving by providing a common factual basis for people differing in perspective. Currently AgriFACTs requires the user to extrapolate Experiment Station results to farm conditions. The next development phase will support map interpretation by forecasting yields across the agricultural landscape.

The Problem

International market dynamics, technological innovations and continental weather systems sweep across Great Plains agriculture; yielding abundance as well as devastation. On occasion, farm policies buffer extreme effects. Generations of farm families struggle with economic depression, drought, easy credit and low commodity prices. Current U.S. farm policy offers farmers the challenge of crop selection. A bit of history may provide appreciation for the opportunities and risks associated with farmers’ choices.

Water and productivity

Variability distinguishes the continental climate of the central Great Plains. Farmers smirk at graphs of “normal” weather patterns, for rarely do two years out of thirty have similar precipitation, radiation and temperature sequences. Rainfall frequently limits productivity in the semi-arid regions of the Great Plains (generally considered west of the 100th meridian). Annual precipitation may meet less than one third of atmospheric evaporative demand in the High Plains.

The wheat-fallow crop system supported farm families in the central High Plains for generations, yielding reliable crops once each two-year cycle. Farm policy controlling acreage to stabilize commodity prices reinforced the wheat-fallow cropping system, by buffering price fluctuations.

While policies supporting wheat-fallow provided stability, the system suffered inefficiencies. Conventional tillage pulverized fragile soil aggregates, leaving soil vulnerable to erosion by wind and water. Water balance studies show the wheat-fallow system utilizes 40% or less of precipitation during a two-year period (Farahani et al., 1998). In contrast, annual cropping can double land productivity, relative to wheat-fallow systems (Anderson et al., 1999a) by increasing the crop transpiration fraction of total evaporation (Peterson et al., 1996). The “Freedom to Farm” policy established in 1996 reduced constraints in crop selection. Many farmers are aggressively exercising crop choice to implement alternatives to the wheat-fallow cropping system.

What crop to plant?

We focus on crop selection, as opposed to in-field crop management, considering the relative opportunities to add value to water-limiting cropping systems. The potential to add value to semi-arid rain-fed grain production is limited by land productivity and by commodity pricing. In the central high Plains, typical returns to land, labor and management are in the order of $100 per hectare or 1ó per square meter (Rosales et al., 1996). Operational control systems impacting field management zones at scales in the order of 1,000 m2 receive consideration due to production and environmental benefits. We hold that decision-support systems impacting crop selection at scales in the order of 1,000,000 m2 warrant consideration due to implications for financial stability of the farm enterprise (Anderson et al., 1999b).

Crops differ in water required during a growing season, in above-ground biomass produced and in the grain fraction of above-ground biomass. These differences arise from multiple factors including heat and humidity during the growing season, leaf and whole-plant stress physiology and composition of harvested seed.

Crop yield can be simply represented as a linear function of available water (effective rainfall, net depletion of stored soil water and any supplemental irrigation). Warm-season grain crops (C4 physiology) can be most productive. Cool-season grain crops (C3 physiology) are intermediate. Productivity of oilseed crops (C3 physiology) appears lowest, though similar to that of cool-season grains when the higher energy content of oil is taken into account.

Cropping systems studies are currently located at multiple locations in the region including: Akron, Stratton, and Sterling, Colorado; Sidney, Nebraska; and Wall, South Dakota. Each research site is conducting long-term crop systems research representative of specific growing environments. For example, at the Akron research site, scientists are evaluating different cropping rotations such as wheat-corn-corn, wheat-corn-proso millet, and wheat-proso millet. (Anderson et al., 1999b) At the Wall, SD research site, successful rotations include wheat-sunflower-proso millet and wheat-field peas-proso millet. (Stymiest, 1997) There is a wide variation of soil properties and climatic characteristics between the sites. For example, annual precipitation can range from 300 mm on the western side to 600 mm on the eastern side of the central Great Plains study region.

Problem summary

Freedom to Farm agricultural policy challenges farmers to select crops, which are most productive and profitable for the land they manage. Cropping system studies at multiple central High Plains locations provide a basis for crop selection. The question arises, which sets of experimental results are most relevant to a given farmer’s growing environment?

The Product

Vision

Our vision is to give farmers and their advisors access to information pertinent to crop selection. Internet-based weather websites provide a working model for real-time geospatial decision-support. These interactive websites integrate mapping, remote-sensing and meteorological information to support weather-related risk management and planning. We’re seeking an analogous decision-support system for cropping systems, offering convenient access to reliable information pertinent to crop selection.

We developed a web-based wide-area (1:250,000 scale) spatial decision support system. Agricultural scientists and GIS specialists integrated results of agroecoregions modeling (USDA SCS, 1983) with ARS cropping systems research to provide farmers with data access. Soil characteristics and weather-related risks affect the suitability of alternative cropping practices for a particular farm location. Though this type of research information is not readily available to farmers, it may help the farmer assess impacts of crop selection for new cropping systems on their farms.

Prototype development

In this project, a coupled-parameter approach was developed to delineate regions of similarity surrounding agricultural experiment stations; defining inference spaces to extrapolate research results on crop rotations and tillage systems to farms in the region. Six different soil and climatic factors were modeled across the central Great Plains landscape at the NRCS National Soil Survey Center in Lincoln, Nebraska.

Two of the modeled surfaces deal with soil properties:

  • Root Zone Available Water-Holding Capacity (RZWHC),
  • Soil Rating for Plant Growth (SRPG, Soil Survey Staff, 2000)

Four climatic characteristic surface models include:

  • Precipitation (PRECIP)
  • Potential Evapotranspiration (PET)
  • Annual Water Balance (AWB).
  • Growing Degree Days (GDD).

These data were considered relevant to cropping system analysis in the central Great Plains. The climate and soil surface models were developed using terrain regression techniques from point data such as weather station and soil characterization sites, as well as from soil geographic databases (USDA SCS, 1994). The figure to the left shows the regional surface model for normal annual precipitation.

Ancillary geospatial data to provide cultural features and context include:

  • analytical hillshade (landform map of the area);
  • county and state boundaries;
  • roads;
  • rivers;
  • towns;
  • cropping systems research sites.

How Agri-FACTs Works

Agri-FACTs is designed to be user friendly. Users can perform two basic operations: data exploration and data extraction. For data exploration, farmers can zoom in to the location of their farms and examine the soil and climate factors for that particular location by turning the AgriFACTs themes on and off. For example, in the figure to the left, the Root Zone Available Water Holding Capacity theme is turned on, allowing the user to see these data for the general area.

For data extraction, the farmer clicks on the map to indicate the location of his farm, as shown in the figure to the right. The program checks all of the soil- and climate-related spatial data layers for the selected location and produces the site-report shown in the following page.

The Agri-FACTs site-report displays the values for six models (PET, AWB, SRPG, PRECIP, RZWHC, and GDD) for the selected location. The report allows the farmer to individually compare each of the soil and climatic values found at the farm location to the soil and climatic values found at each of the five research sites. The farmer can then determine which research station(s) has the most similar soil and climatic characteristics to their farm. The report contains hyperlinks to each of the experiment stations, allowing the farmer to go directly to the research site’s web pages and examine the cropping system research results. It is at this level that Agri-FACTs facilitates decision support regarding new cropping systems that the producer may elect to consider for his farm to replace the wheat-fallow system most commonly used. The farmer can then take this information to his extension agent or district conservationist for a more explicit analysis.

Reactions to the product

Responses from farmers, crop consultants and agricultural scientists come from workshops, symposia and the Internet. A typical initial reaction is quiet observation. Maps of soil and weather characteristics can grasp and retain a dryland farmer’s attention. More than one extension agent focused on into the RZWHC theme and verified the delineation of fine- and coarse-textured soils for his region. Comparisons of rainfall and evaporative demand are of particular interest.

Frequently, viewers request further information. Requests for factual detail included

  • monthly precipitation charts, with measures of uncertainty,
  • thematic maps of growing degree days, computed by month, and
  • thematic maps of highly erodible lands.

Farmers, scientists and administrators also wanted more agronomic interpretation of the geographic information. Queries included:

  • What is the risk of crop failure for a given location?
  • For what region are crop yields likely to exceed break-even 80% of all cropping seasons?
  • Are 90-day grain sorghum varieties more likely to avoid frost damage than 100 day varieties in this locality?

AgriFACTs also stimulated queries regarding climatic processes:

  • Do ENSO-related patterns affect land productivity?
  • What accounts for greater rainfall to the east?

GIS developers wanted to learn about new developments, other models for thematic maps considered for inclusion in AgriFACTs, and suggestions for developing similar applications.

AgriFACTs viewers also responded with creative action. More than one farmer has contributed personal historic rainfall observations for inclusion in the database. An agronomist analyzed pan evaporation data throughout the Great Plains, providing a model suitable for a thematic map. A research administrator commissioned a spatial analysis identifying agroecozones surrounding each experiment station within his administrative region. He wanted to compare soils to assess applicability of experimental results. A groundwater administrator commented that map products similar to AgriFACTs elevate the plane of discussion by providing a common factual basis for people with strong differences in perspective.

Product summary

AgriFACTs provides web-accessible maps of soil and weather characteristics of the central High Plains. Farmers may use these maps to identify cropping system studies most relevant to their farm conditions. Farmers, their advisors and agricultural scientists responded to AgriFACTs with skeptical interest; valuing accuracy and requesting more detailed information. The maps provide common ground for dialogue and a platform stimulating creative action.

The Process (on-going)

AgriFACTs requires that farmers infer crop productivity and agronomic risk on their farms from cropping systems studies conducted some distance from their farms. Spatial decision support suggests applying agronomic principles across the landscape, e.g., linking expected yields with soil and weather conditions expected at the farm site. This suggests developing a capability for geospatial yield forecasting. A brief discussion of relevant development processes follows.

The process of interpreting spatial information can be idealized as a sequence of creative processes involving sensor design, signal processing, data interpretation and application.

innovation signals data information utility

As with most creative processes, the reality can be a bit messier. Assumptions, approximations and errors along the path can limit ultimate utility.

Systems science offers the objective function for quantifying optimal design and management of a defined system. An objective function can be presented in the form of state-space equations (Rao, 2001).

Here x is a n-vector (matrix of nĚ1) of state variables (e.g. soil water and nutrient status, crop canopy, growth stage, etc.), u is a r-vector of inputs (weather, crop cultural practices, prices, etc.) and y is a p-vector of outputs representing value-based measures of system output (productivity, profitability, environmental impacts, etc.). A, B, C and D are matrices of dimension nĚn, nĚr, pĚn and pĚr, respectively. These coefficients are intended to describe system properties and relationships. Linear analysis provides powerful tools to handle complex dynamic systems, when they are well defined.

Simplified algorithms can be calibrated to approximate crop growth responses to environmental variation. Three general approaches are based on

  • correlation of historic crop yields with agroclimatic factors,
  • productivity arising from light incepted by the crop canopy, and
  • linkage of assimilated CO2 with crop transpiration.

Henry Wallace, U.S. Secretary of Agriculture (1930’s) developed an agroclimatic model of productivity, which may provide utility for decision-support. He related yield deviations from normal to weather deviations from normal. The original equations provided best estimates of corn yield in central Pennsylvania than contemporary equations, and better than two process-oriented models (Norman, 1993). This type of algorithm may provide a first-order approximation of yield across the agricultural landscape.

Remote sensing of crop canopy status provides information about productivity, due to its relationship to light interception (Kiniry et al., 1998). Regional crop yield forecasts, calibrated to USDA crop estimates, use imagery derived from Advance Very High Resolution Radiometry imagery (Kastens et al., 1999). Archived composite imagery is available on a global, two-week interval basis for historical analysis (Reed et al., 1996).

Transpiration-based productivity forecasts may be suitable for water-limited regions (Tanner and Sinclair, 1983). When complemented by water balance analyses, other hydrologic processes may be evaluated as well. As a first approximation, yield response to available water, combined with historic rainfall can indicate frequency distributions of expected yields for a given crop at a given site.

Currently, work at K-State NWREC is underway to evaluate the predictive accuracy of agroclimatic, radiation- and transpiration-based models of crop productivity. Implemented as state-space equations, these algorithms are capable of providing web-based access to a dynamic agroclimatic GIS. Further, we’re investigating the utility of implementing a geospatial water balance system simulation in daily time step increments. The envisioned system would utilize NEXRAD precipitation images and require calibration to fit USGS stream-flow observations. Such a system could be suitable to quantify agronomic risk factors, water quality impacts of land management and groundwater recharge dynamics. Whether these decision support tools can provide information with sufficient accuracy for utility remains to be proven.

Process summary

Applying cropping systems principles across the agricultural landscape implies yield forecasting with significant spatial resolution. Crop productivity can be related to agroclimatic factors, light interception, and/or crop transpiration rates. Implementing accurate and efficient algorithms as state-space equations may provide users with powerful analytic tools to evaluate crop selection in a high-risk environment.

Acknowledgements

We acknowledge the contributions of Drs. Randall Andersen, Ray Sinclair and Sharon Waltman in the development and application of the database supporting AgriFACTs.

Biography

Vern Thomas has a Bachelors Degree in Landscape Architecture from the University of Idaho (1986) and a Masters Degree in Remote Sensing and GIS from Colorado State University (1994). He has been a GIS Analyst and Applications Developer since January 1996 and has worked on many different applications with NRCS (MLRA Revision, Conservation Area Resource Assessment Analysis), USFS (Arapaho-Roosevelt National Forest Management Plan SDSS), and ARS (Agroecoregions). Presently, Vern is working as a contractor specialist in Remote Sensing and GIS for the United States Forest Service, Forest Health Protection, Forest Health Technology Enterprise Team developing new methodologies to furnish technical expertise in the detection, monitoring, and evaluation of forest health related concerns.

Bill Waltman received his B.S. (1978), M.S. (1981), and Ph.D. (1985) in Agronomy from The Pennsylvania State University. Upon graduation from Penn State, Bill directed the Soil Characterization Laboratory at Cornell University and conducted an extension program in Soil Interpretations and Land Use. In 1991, he joined the former USDA Soil Conservation Service and was a Research Soil Scientist and Regional GIS Specialist, stationed in Lincoln, Nebraska. Bill works as a Research Coordinator for the Nebraska Research Initiative on Geospatial Decision Support Systems. Bill's current research program focuses on the development of a drought decision support system for agriculture and the geospatial modeling of soil climate regimes.

Rob Aiken received his B.S. (1977) and M.S. (1988), from the University of Nebraska and Ph.D. (1991) from Michigan State University. Trained in environmental biophysics and crop simulation modeling, his research focus is improved water use in rain-fed semi-arid cropping systems. Emerging applications include risk assessment tools, which combine spatial databases, remote sensing and simple crop productivity models for wide-area crop yield forecasting and climate-risk management.

References

Anderson R.L., R.M. Aiken, H.R. Sinclair, S.W. Waltman and W.J. Waltman. (1999b) Identifying Agroecozones in the Central Great Plains. Proceedings of the Fourth International Conference on Precision Agriculture. pp 1347-1353.

Anderson, R.L., R.A. Bowman, D.C. Nielsen, M.F. Vigil, R.M. Aiken and J.G. Benjamin. (1999a) Alternative crop rotations for the central Great Plains. J. Prod. Agric. 12:95-99.

Farahani, H.J., G.A. Peterson and D.G. Westfall. (1998). Dryland cropping intensification—A fundamental solution to efficient use of precipitation. Advances in Agronomy. 64:197-223.

Greb, B.W. (1983). Water conservation: Central Great Plains. P. 57-73. In H.E. Dregne and W.O. Willis (ed.) Dryland Agriculture. Agron. Monogr. 23. ASA, CSSA, and SSSA, Madison, WI.

Kastens, D.L.A., K.P. Price, E.A. Martinko and T.L. Kansens. (1999). Estimating pre-harvest wheat yields from time-series analysis of remotely sensed data. Proc., Ann. Conv., Amerc. Soc. Photo. Eng and Remote Sensing.

Kiniry, J.R., C.A. Jones, J.C. O’Toole, R. Blanchet, M. Cabelguenne and D. A. Spanel. (1989) Radiation-use efficiency in biomass accumulation prior to grain-filling for five grain-crop species. Field Crops Res. 20:51-64.

Norman, J.M. (1993) Scaling processes between leaf and canopy levels. In J.R. Ehleringer and C.B. Field. Scaling physiological processes. Academic Press, Inc. San Diego. pp. 41-76.

Peterson, G.A., A.J. Schlegel, D.L. Tanaka and O.R. Jones. (1996). Precipitation use efficiency as affected by cropping and tillage systems. J. Prod. Agric. 9:180-186.

Rao, G.P. (2001) Elements of Control Systems. Encyclopedia of Life Support Systems. http://www.eolss.com/

Reed, B.C., T.R. Loveland and L.L. Tieszen. (1996) An approach for using AVHRR data to monitor U.S. Great Plains Grasslands. Geocarto Int. 11(3):13-22

Rosales, M., J. Saunders, M. Sucik, G. Uhler and D. Nitchie. (1996) Alternative Rotations to Wheat-Fallow. USDA/NRCS Soil Quality Team. Akron, CO.

Soil Survey Staff. (2000) Soil Ratings for Plant Growth—A System for Arraying Soils According to their Inherent Productivity and Suitability for Crops. USDA/NRCS National Soil Survey Center, Lincoln, NE.

Stymiest, C.E. (1997) Annual progress report. South Dakota State Univ. Plant Science Pamphlet No. 90.

Tanner, C.B. and T.R. Sinclair. (1983) Efficient water use in crop production: Research or re-search? In Limitations to efficient water use in crop production. ASA. Madison, WI.

USDA Soil Conservation Service. (1983) Land Resource Regions and Major Land Resource Areas of the United States. Agric. Handbook No. 296, U.S. Gov’t Printing Office, Washington, D.C.

USDA Soil Conservation Service. (1994). State Soil Geographic Database (STATSGO), User’s Guide. Miscellaneous Publication No. 1492, National Soil Survey Center, Lincoln, NE.

Top Of PageNext Page